From: "Michael S. Tsirkin" <mst@redhat.com>
To: Matti Moell <matti.moell@oss.qualcomm.com>
Cc: Peter Hilber <peter.hilber@oss.qualcomm.com>,
Cornelia Huck <cohuck@redhat.com>,
Demi Marie Obenour <demiobenour@gmail.com>,
Manos Pitsidianakis <manos.pitsidianakis@linaro.org>,
Parav Pandit <parav@nvidia.com>,
Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>,
"virtio-comment@lists.linux.dev" <virtio-comment@lists.linux.dev>,
Jan Kiszka <jan.kiszka@siemens.com>,
Stefan Hajnoczi <stefanha@redhat.com>,
Trilok Soni <trilok.soni@oss.qualcomm.com>
Subject: Re: [PATCH v2] content: Reserve ID 52 for QPSFI device
Date: Wed, 4 Nov 2026 09:46:39 -0500 [thread overview]
Message-ID: <20261104094014-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <4222d5ec-e44f-46a1-8a14-a2f3fb6d84d1@oss.qualcomm.com>
On Wed, Nov 04, 2026 at 03:34:00PM +0100, Matti Moell wrote:
> On 28/10/2026 15:22, Peter Hilber wrote:
> > On Mon, Oct 26, 2026 at 03:31:16PM +0100, Cornelia Huck wrote:
> > >
> > > ...this would also still somewhat hold (no need to reserve an ID for
> > > everything, but if you're serious, get a vendor-or-purpose prefix.)
> > > I'd still want to encourage everyone creating a serious device to go and
> > > try to standardize it.
> > >
> >
> > I am wondering how these namespaces would be managed and used. Would
> > introducing a vendor/purpose prefix require an approval process in the
> > Virtio TC? Would new standardized devices have a "standard" prefix when
> > using ID 52?
>
> I would suggest something like this
> <PURPOSE>/<UUID>
>
> Where PURPOSE is just some well known 32 bit number with:
> - 0x..00: must be ignored (similar to the device IDs)
> - 0x..01: test and experimentation
> - 0x..02: Vendor space, any vendor can use it
> - 0x..03 - 0x..FF: Reserved for future use
>
> Then the UUID can be any completely random within the purpose. Not "4" but a
> meaningfully generated UUIDv4.
>
> OR alternatively we could create a QPSFI or maybe Qualcomm purpose instead
> of using a generic vendor space, for example 0x..02 could just be used for
> QPSFI. I don't think it makes a huge difference.
>
> > Could a device UUID transparently transition from testing to
> > vendor/purpose, so that drivers would only match on the UUID? Could the
> > namespaces be seen as a quality gate so that testing-prefix devices may
> > be refused in some circumstances?
> I don't think that should be thing, once something is properly specified it
> can just use a normal device ID.
Ugh. Let's keep it simple. Reserve ID 52 as vendor specific, qualcom
can do what qualcom wants with a qualcom subsystem vendor id,
or just you know, use its own device/vendor ID.
>
> On another note. We are here because there just aren't enough device IDs
> available
Nope, I don't think we want to go too far to accomodate devices with no
plans to be part of the virtio spec. That's why we are here,
the lack of IDs we'll fix.
> and while looking at this we figured out that the qemu docs[1]
> actually claim that there are 175 device IDs reserved for VIRTIO and not 63
> as written in the spec. Who's wrong here?
>
> Maybe the solution lies in finding ways to expand the number of available
> IDs.
>
> Cheers,
>
> Matti
>
> [1]: https://www.qemu.org/docs/master/specs/pci-ids.html
--
MST
next prev parent reply other threads:[~2026-11-04 14:46 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-21 18:00 [PATCH v2] content: Reserve ID 52 for QPSFI device Peter Hilber
2026-10-21 19:25 ` Demi Marie Obenour
2026-10-21 19:50 ` Michael S. Tsirkin
2026-10-22 4:35 ` Cornelia Huck
2026-10-22 7:07 ` Demi Marie Obenour
2026-10-22 7:24 ` Michael S. Tsirkin
2026-10-22 8:41 ` Manos Pitsidianakis
2026-10-22 13:19 ` Michael S. Tsirkin
2026-10-22 14:07 ` Dmitry Baryshkov
2026-10-22 15:05 ` Cornelia Huck
2026-10-22 15:35 ` Parav Pandit
2026-10-22 15:41 ` Michael S. Tsirkin
2026-10-22 15:49 ` Parav Pandit
2026-10-22 16:02 ` Michael S. Tsirkin
2026-10-22 16:20 ` Parav Pandit
2026-10-22 16:29 ` Michael S. Tsirkin
2026-10-22 22:39 ` Peter Hilber
2026-10-22 23:28 ` Michael S. Tsirkin
2026-10-23 3:32 ` Parav Pandit
2026-10-23 5:37 ` Manos Pitsidianakis
2026-10-23 5:40 ` Demi Marie Obenour
2026-10-23 20:56 ` Peter Hilber
2026-10-23 23:47 ` Michael S. Tsirkin
2026-10-26 14:31 ` Cornelia Huck
2026-10-28 14:22 ` Peter Hilber
2026-11-04 14:34 ` Matti Moell
2026-11-04 14:46 ` Michael S. Tsirkin [this message]
2026-10-22 15:44 ` Michael S. Tsirkin
2026-10-22 14:50 ` Zelong Ren
2026-10-21 19:41 ` Michael S. Tsirkin
2026-10-22 13:56 ` Dmitry Baryshkov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20261104094014-mutt-send-email-mst@kernel.org \
--to=mst@redhat.com \
--cc=cohuck@redhat.com \
--cc=demiobenour@gmail.com \
--cc=dmitry.baryshkov@oss.qualcomm.com \
--cc=jan.kiszka@siemens.com \
--cc=manos.pitsidianakis@linaro.org \
--cc=matti.moell@oss.qualcomm.com \
--cc=parav@nvidia.com \
--cc=peter.hilber@oss.qualcomm.com \
--cc=stefanha@redhat.com \
--cc=trilok.soni@oss.qualcomm.com \
--cc=virtio-comment@lists.linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox